Facebook’s Popularity Plummets in Mideast. Users Suspect Censorship During Israel-Palestinian Conflict.
Facebook is grappling with a reputation crisis in the Middle East, with plummeting approval rates and advertising sales in Arab countries, according to leaked documents obtained by NBC News.
The shift corresponds with the widespread belief by pro-Palestinian and free speech activists that the social media company has been disproportionately silencing Palestinian voices on its apps – which include Facebook, Instagram and WhatsApp – during this month’s Israel-Hamas conflict. Examples include the deletion of hundreds of posts condemning the eviction of Palestinians from the Sheikh Jarrah neighborhood of Jerusalem, the suspension of activist accounts and the temporary blocking of a hashtag relating to one of Islam’s holiest mosques. Facebook said these were technical glitches.
Instagram has taken the greatest reputational hit, according to a presentation authored by a Dubai-based Facebook employee that was leaked to NBC News, with its approval ratings among users falling to a historical low.
The social media company regularly polls users of Facebook and Instagram about how much they believe the company cares about them. Facebook converts the results into a ‘Cares About Users’ metric which acts as a bellwether for the apps’ popularity. Since the start of the latest Israel-Hamas conflict, the metric among Instagram users in Facebook’s Middle East and North Africa region is at its lowest in history, and fell almost 5 percentage points in a week, according to the research. “The biggest changes came from Qatar, Jordan, Palestine and Saudi Arabia,” the presentation states.
Instagram’s score measuring whether users think the app is good for the world, referred to as ‘Good For World,’ has also dropped in the region to its lowest level after losing more than 5 percentage points in a week.
There was also a dip — although not as precipitous — in Facebook’s ‘Cares About Users’ score in the Middle East, driven mainly by Egypt, Iraq, Morocco and Jordan.
The low approval ratings have been compounded by a campaign by pro-Palestinian and free speech activists to target Facebook with 1-star reviews on the Apple and Google app stores. The campaign tanked Facebook’s average rating from above 4 out of 5 stars on both app stores to 2.2 on the App Store and 2.3 on Google Play as of Wednesday. According to leaked internal posts, the issue has been categorized internally as a “severity 1” problem for Facebook, which is the second highest priority issue after a “severity 0” incident, which is reserved for when the website is down.
“Users are feeling that they are being censored, getting limited distribution, and ultimately silenced,” one senior software engineer said in a post on Facebook’s internal message board. “As a result, our users have started protesting by leaving 1 star reviews.”
Internal documents connect the reputational damage to a decline in advertising sales in the Middle East. According to the leaked presentation, Facebook’s ad sales in the United Arab Emirates, Egypt, Morocco, Saudi Arabia, Kuwait, Qatar and Iraq dropped at least 12 percent in the 10 days after May 7.
“In addition to the negative sentiment towards Facebook in MENA now, the regression could also be attributed to the overall charged environment where some brands might find it insensitive to advertise or won’t be getting the usual ROI when spending their money,” Facebook noted in a presentation deck on the topic, referring to the return on investment.
In spite of the widespread perception of disproportionate silencing of Palestinian voices, Facebook has been unable to identify any “ongoing systemic issues” with its automated content removal tools or human content moderators, according to a post to Facebook’s internal message board by the company’s risk and response team.
Facebook spokesperson Andy Stone acknowledged in an interview there had been “several issues” that affected people’s ability to share on the company’s apps. “While we have fixed them, they should never have happened in the first place and we’re sorry to anyone who felt they couldn’t bring attention to important events, or who felt this was a deliberate suppression of their voice. This was never our intention – nor do we ever want to silence a particular community or point of view.”
Stone declined to comment on the leaked material indicating reputation and ad sales damage in the Middle East but did not dispute its contents.
The Middle East is not a huge market in terms of Facebook’s overall advertising revenue, which topped $84 billion in 2020. According to research by the Dubai-based digital advertising company Futuretech, the social media company generates between $800 million and $1 billion in annual advertising revenue in the region. But it is a key growth market at a time when user growth in some of the larger advertising markets such as the United States and Europe has stalled, experts say.
“It might be a smaller market. But Facebook is still growing aggressively in the region. So it’s strategically important,” Futuretech CEO Boye Balogun said. “They are also protecting themselves from the likes of TikTok that have launched in the region and are growing very quickly.”
The same leaked presentation highlighted a surge in Israeli users reporting problematic content on Instagram, first reported by BuzzFeed News, making Israel the top country ranked globally for reporting content for ties to dangerous organizations and individuals, a category that covers terrorist propaganda, between April and May.
There was also an increase in reports from users in Israel of content for violating Facebook’s rules against hate speech and incitement of violence. Between May 8 and May 18, users in Israel reported 494,463 cases of hate speech while Palestinian users reported 58,618 cases.
“Israel has hacked the system and knows how to pressure Facebook to take stuff down,” said Ashraf Zeitoon, who worked as Facebook’s head of policy in the Middle East and North Africa between 2014 and 2017. “Palestinians don’t have the capacity, experience and resources to report hate speech by Israeli citizens in Hebrew.”
The U.S., Israel and the European Union have designated Hamas as a terrorist organization. But pro-Palestinian civil society groups including 7amleh and Access Now as well as some Facebook employees are concerned that nonviolating content is being marked as terrorist propaganda and hate speech. They believe that Israel is flooding Facebook with reports of violations in a way that disproportionately removes Palestinian voices.
“The Israeli government is spending millions on digital tools and campaigns targeting social media content. But there are only fragmented efforts from the Palestinian side,” said Mona Shtaya from 7amleh, a nonprofit that focuses on Palestinians’ digital rights.
Shtaya notes that the Israeli government’s cyber unit makes thousands of requests each year to have content taken down from social media sites, including Facebook and Instagram. During the first 10 days of May, as Israeli-Palestinian tensions were rising, the Israeli government asked social media companies to delete more than 1,010 pieces of content. More than half (598) of the requests were made to Facebook and the Israeli government said that the social media company removed 48 percent of them. Israel also funds a program that pays students to post and report content on social media in what is described as “online public diplomacy.”
Shtaya also points to apps such as Act-IL, developed by former Israeli intelligence officers, where volunteers coordinate campaigns to mass report items of content and boost other items by liking and sharing them. Act-IL has been promoted by the Israeli government, but the company denies any formal relationship and in 2018 told BuzzFeed News that it was a “grassroots initiative” that fights antisemitism and incitement to terrorism and violence.
Free speech and human rights groups have for years accused Facebook of censoring Palestinian voices, including activists and journalists.
Florida hits Big Tech with a first-of-its-kind ‘deplatforming’ law
In November 2016, members of Palestinian solidarity groups including the Jewish Voice for Peace, the Institute for Middle East Understanding and the Palestinian Business Committee for Peace and Reform met with Facebook employees at the company’s Menlo Park, California, headquarters to discuss concerns that Palestinian activists were being silenced and allegations that Israel was trying to game the system using its network of paid students.
According to notes taken at the meeting, reviewed by NBC News, the Palestinian solidarity groups had a key takeaway: “It doesn’t feel like Facebook is taking us or the issue seriously enough. Facebook continues to defend problematic policies and we need to continue educating them about how these policies harm their stated goals of ‘promoting free expression’ and ‘making sure all users are treated the same.’” Facebook spokesperson Stone declined to comment on the 2016 meeting.
But pressure on Facebook has intensified this month as reports of such censorship collected by digital rights groups including Access Now and 7amleh have skyrocketed. They point to hundreds of examples, including Instagram removing posts and blocking hashtags about Al Aqsa, one of Islam’s holiest mosques, because its content moderation system mistakenly connected the name to a terrorist organization.
Facebook has also been attempting to crack down on extremist Israeli content, including more than a hundred WhatsApp groups — identified by The New York Times and FakeReporter, an Israeli watchdog that studies misinformation — where members call for and plan violent attacks on Palestinians.
A WhatsApp spokeswoman confirmed that it had removed some of the accounts of people in the groups and that, although it cannot read the encrypted messages sent through its service, it takes action to “ban accounts that we believe may be involved in causing imminent harm.”
This week, concerns were elevated in Washington, D.C., by Rep. Rashida Tlaib, D-Mich., who called for Facebook and other social media companies to stop censoring Palestinian political speech on their platforms.
“It is not enough to continue to blame technical errors and automated systems and algorithms,” she wrote in a letter to Facebook, Twitter and TikTok.
Internally, the leaked posts show that a group of about 30 Facebook employees have volunteered their time to file what Facebook refers to as “Oops tickets,” a system that allows employees to flag problems to prioritize, connected to issues related to the Israeli-Palestinian conflict, and many of these have been about pro-Palestinian content believed to have been removed by mistake.
Facebook has also set up an Integrity Product Operations Center, comprising native Arabic and Hebrew speakers and subject matter experts from teams including threat intelligence, data science, operations, research, policy and legal to handle issues related to the conflict, including restoring content that has been mistakenly removed.
Facebook has regular meetings with senior members of the Israeli government, including one May 13 with Israeli Justice Minister Benny Gantz, during which he pressured Facebook to be more proactive in taking action against “extremist elements that are seeking to do damage to our country,” according a statement released by his office.
But on May 13, senior Facebook executives including Nick Clegg, vice president for global affairs and communications, Joel Kaplan, vice president of global public policy, and MENA policy chief Azzam Alameddin met virtually with Palestinian Prime Minister Mohammad Shtayyeh — the first such meeting to have taken place. Previously members of the Palestinian government and civil society groups had met with Facebook employees, Facebook spokesman Stone said. However Palestinian politicians and civil society groups have never before had access to such senior executives, 7amleh’s Mona Shtaya said.
“Facebook started by saying they felt sorry for Palestinians and what was happening on the ground,” said Shtaya, who attended the meeting with the Palestinian prime minister representing civil society. “The feeling from the meeting was that Facebook came to prove to people that they aren’t biased and that they are not only sitting with the Israeli government but also the Palestinian government.”
The group spent some time discussing words regularly used by Palestinians, including “martyr” and “resistance” that were being labeled by Facebook as hate speech or incitement to violence.
“They showed openness, they acknowledged some mistakes, and promised to take these points into consideration,” said Shtaya, who said that the meeting went “really well.”
Stone said that the meetings with Gantz and Shtayyeh were “an effort to ensure that all parties are aware of steps the company has taken, and will continue to take, to keep the platform safe.”
On Wednesday, Facebook representatives met with pro-Palestinian and free speech civil society groups, including the Electronic Frontier Foundation and Access Now, to discuss the same issues.
Access Now’s Marwa Fatafta said the meeting was arranged after they sent an email to Facebook CEO Mark Zuckerberg, hoping to sit down with executive leadership. Instead they met with Facebook’s human rights policy team and asked for a public audit of the company’s content policies related to the Israeli-Palestinian issue.
Fatafta came away thinking that the Facebook attendees were well-intentioned, but lacked power to effect meaningful change.
“I am always skeptical whenever their human rights team wants to meet with us. We know they have not so much influence within the organization,” Fatafta said. “I always feel their job is to manage down civil society and their concerns. But they don’t do much when it comes to actual structural changes.”
Stone said that the meeting was an “important opportunity to hear their concerns directly” and that “we share an interest in making sure Facebook remains a place for Palestinians and others around the world to discuss the issues that matter to them.”